Matches in SemOpenAlex for { <https://semopenalex.org/work/W2913666739> ?p ?o ?g. }
- W2913666739 abstract "Random survival forest and survival trees are popular models in statistics and machine learning. However, there is a lack of general understanding regarding consistency, splitting rules and influence of the censoring mechanism. In this paper, we investigate the statistical properties of existing methods from several interesting perspectives. First, we show that traditional splitting rules with censored outcomes rely on a biased estimation of the within-node failure distribution. To exactly quantify this bias, we develop a concentration bound of the within-node estimation based on non i.i.d. samples and apply it to the entire forest. Second, we analyze the entanglement between the failure and censoring distributions caused by univariate splits, and show that without correcting the bias at an internal node, survival tree and forest models can still enjoy consistency under suitable conditions. In particular, we demonstrate this property under two cases: a finite-dimensional case where the splitting variables and cutting points are chosen randomly, and a high-dimensional case where the covariates are weakly correlated. Our results can also degenerate into an independent covariate setting, which is commonly used in the random forest literature for high-dimensional sparse models. However, it may not be avoidable that the convergence rate depends on the total number of variables in the failure and censoring distributions. Third, we propose a new splitting rule that compares bias-corrected cumulative hazard functions at each internal node. We show that the rate of consistency of this new model depends only on the number of failure variables, which improves from non-bias-corrected versions. We perform simulation studies to confirm that this can substantially benefit the prediction error." @default.
- W2913666739 created "2019-02-21" @default.
- W2913666739 creator A5010394308 @default.
- W2913666739 creator A5045963637 @default.
- W2913666739 creator A5049569229 @default.
- W2913666739 creator A5076762447 @default.
- W2913666739 date "2017-07-30" @default.
- W2913666739 modified "2023-09-23" @default.
- W2913666739 title "Consistency of survival tree and forest models: splitting bias and correction" @default.
- W2913666739 cites W128574488 @default.
- W2913666739 cites W13188192 @default.
- W2913666739 cites W150987698 @default.
- W2913666739 cites W1522921740 @default.
- W2913666739 cites W1971163323 @default.
- W2913666739 cites W1978391225 @default.
- W2913666739 cites W1979300931 @default.
- W2913666739 cites W1980485115 @default.
- W2913666739 cites W1982251463 @default.
- W2913666739 cites W2010029425 @default.
- W2913666739 cites W2011186625 @default.
- W2913666739 cites W2014164843 @default.
- W2913666739 cites W2016681503 @default.
- W2913666739 cites W2019412363 @default.
- W2913666739 cites W2026925128 @default.
- W2913666739 cites W2035597599 @default.
- W2913666739 cites W2036553170 @default.
- W2913666739 cites W2040311524 @default.
- W2913666739 cites W2047081748 @default.
- W2913666739 cites W2056132907 @default.
- W2913666739 cites W2057968703 @default.
- W2913666739 cites W2059989526 @default.
- W2913666739 cites W2069737045 @default.
- W2913666739 cites W2073672062 @default.
- W2913666739 cites W2078399076 @default.
- W2913666739 cites W2080791146 @default.
- W2913666739 cites W2083274139 @default.
- W2913666739 cites W2084644495 @default.
- W2913666739 cites W2085107973 @default.
- W2913666739 cites W2093729382 @default.
- W2913666739 cites W2099990492 @default.
- W2913666739 cites W2108549943 @default.
- W2913666739 cites W2108704825 @default.
- W2913666739 cites W2117308775 @default.
- W2913666739 cites W2120985611 @default.
- W2913666739 cites W2122858723 @default.
- W2913666739 cites W2123998733 @default.
- W2913666739 cites W2127611910 @default.
- W2913666739 cites W2148414029 @default.
- W2913666739 cites W2149446614 @default.
- W2913666739 cites W2208550830 @default.
- W2913666739 cites W2214916291 @default.
- W2913666739 cites W2216946510 @default.
- W2913666739 cites W2314575834 @default.
- W2913666739 cites W2325180252 @default.
- W2913666739 cites W2345898006 @default.
- W2913666739 cites W2474271774 @default.
- W2913666739 cites W2477400917 @default.
- W2913666739 cites W2561265960 @default.
- W2913666739 cites W2564036198 @default.
- W2913666739 cites W2801653770 @default.
- W2913666739 cites W2911964244 @default.
- W2913666739 cites W2951703004 @default.
- W2913666739 cites W2963820464 @default.
- W2913666739 cites W2993854555 @default.
- W2913666739 cites W3102027041 @default.
- W2913666739 cites W339716 @default.
- W2913666739 cites W41000882 @default.
- W2913666739 cites W2108719358 @default.
- W2913666739 cites W2112740000 @default.
- W2913666739 doi "https://doi.org/10.48550/arxiv.1707.09631" @default.
- W2913666739 hasPublicationYear "2017" @default.
- W2913666739 type Work @default.
- W2913666739 sameAs 2913666739 @default.
- W2913666739 citedByCount "7" @default.
- W2913666739 countsByYear W29136667392018 @default.
- W2913666739 countsByYear W29136667392019 @default.
- W2913666739 crossrefType "posted-content" @default.
- W2913666739 hasAuthorship W2913666739A5010394308 @default.
- W2913666739 hasAuthorship W2913666739A5045963637 @default.
- W2913666739 hasAuthorship W2913666739A5049569229 @default.
- W2913666739 hasAuthorship W2913666739A5076762447 @default.
- W2913666739 hasBestOaLocation W29136667391 @default.
- W2913666739 hasConcept C105795698 @default.
- W2913666739 hasConcept C118615104 @default.
- W2913666739 hasConcept C119043178 @default.
- W2913666739 hasConcept C127413603 @default.
- W2913666739 hasConcept C137668524 @default.
- W2913666739 hasConcept C149782125 @default.
- W2913666739 hasConcept C154945302 @default.
- W2913666739 hasConcept C161584116 @default.
- W2913666739 hasConcept C169258074 @default.
- W2913666739 hasConcept C185429906 @default.
- W2913666739 hasConcept C199163554 @default.
- W2913666739 hasConcept C2776436953 @default.
- W2913666739 hasConcept C31388003 @default.
- W2913666739 hasConcept C33923547 @default.
- W2913666739 hasConcept C41008148 @default.
- W2913666739 hasConcept C62611344 @default.
- W2913666739 hasConcept C66938386 @default.
- W2913666739 hasConceptScore W2913666739C105795698 @default.